Lessons Learned From Nitrogen Doping at JLab

Ari D. Palczewski, SRF Scientist Jefferson Lab, USA

9/14/2015

Exploration of Surface Resistance with Varied Interstitial Atom Diffusion on Niobium Cavity Surfaces

MOBA07

What to remember/what I learned 1) It now seem clear, Nitrogen doping i.e. getting nitrogen into

cavity at 800C is the easy part. 2) Controlling the test environments: magnetic field, cooling rate,

and test hardware is the hard part. Environmental monitoring instrumentation is your friend.

3) Q vs Eacc is not enough, Q vs Eacc vs T in a controlled

environment with “surface resistance decomposition” is a must (all data that follows is 1.3GHz @ 2.0K unless noted).

4) What is the nitrogen doing? Where is it going?

Cavity tests to date – 2.0K

9 cell mini production and re-doping second run

Cavity tests to date – 2.0K

Single cell systematic doping

Cavity tests to date – 2.0K

Single cell large grain incremental EP – Dhakal et al, IPAC 2015 WEPWI009

Cavity tests to date – 2.0K

Large grain vs fine grain –systematic removal

Furnace doping - absorption rate

See Report for JLab High Q0 R&D for LCLS-II - FY14,JLab tech note JLAB-TN-15-008

Test environment – accidental uniform cooling D8

Test environment – accidental uniform cooling D8

Temperature sensors

Test environment – accidental uniform cooling D8

Temperature sensors

Test environment - magnetic field drift

Report for JLab High Q0 R&D for LCLS-II - FY14 JLab tech note JLAB-TN-15-008

Test environment - magnetic field drift

Report for JLab High Q0 R&D for LCLS-II - FY14 JLab tech note JLAB-TN-15-008

Surface decomposition is essential - Qo vs. Eacc vs. T

All single cell cavities, all with now know good doping, all should easily have Q0 > 3.5e10 @ 16MV/m

Surface decomposition is essential - Qo vs. Eacc vs. T

All single cell cavities, all with now know good doping, all should easily have Q0 > 3.5e10 @ 16MV/m

Uniform cooling, 7 mGauss field and BCP pitted cavity

Surface decomposition is essential - Qo vs. Eacc vs. T

All single cell cavities, all with now know good doping, all should easily have Q0 > 3.5e10 @ 16MV/m

Uniform cooling, 5 mGauss field

Surface decomposition is essential - Qo vs. Eacc vs. T

All single cell cavities, all with now know good doping, all should easily have Q0 > 3.5e10 @ 16MV/m

Stratified cooling 2 mGauss field

Surface decomposition is essential - Qo vs. Eacc vs. T

All single cell cavities, all with now know good doping, all should easily have Q0 > 3.5e10 @ 16MV/m

Uniform cooling, 5 mGuass field Stratified cooling 2 mGuass field

1.5K

2.1K

Surface decomposition is essential - Qo vs. Eacc vs. T

All single cell cavities, all with now know good doping, all should easily have Q0 > 3.5e10 @ 16MV/m

Uniform cooling, 7 mGuass field and BCP pitted cavity Uniform cooling, 5 mGuass field Stratified cooling 2 mGuass field

Mostly environmental

Mostly doping

,

G2 – Large grain incremental EP

Bas

elin

e

N2A

30 +

EP1

0

+ E

P5

+ 9

0C 1

6hr

+ E

P5

+ 1

20C

16h

r

+ E

P10

+ 1

20C

16h

r

+ E

P5

+ 1

20C

16h

r

Courtesy of R.L. Geng IPAC 2015 WEPWI013 for Q vs Eacc

Quench analysis on 9 cell cavities

Palczewski et al. – IPAC2015 WEPWI019 QUENCH STUDIES OF SIX HIGH TEMPERATURE NITROGEN DOPED 9 CELL CAVITIES FOR USE IN THE LCLS-II BASELINE PROTOTYPE CRYOMODULE AT JEFFERSON LABORATORY

N20A30 +EP 16μm Average quench field – 16.8MV/m N2A6 + EP 5 μm Average quench field – 23MV/m lower Q0 @ 16MV/m With lower N2?

LCLS-II prototype results before and after HV welding

Average Q0 is lower after HV welding by ~10% or 1.25nΩ

Current data suggest environmental effects are the major cause of Q0 degradation after HV welding. Palczewski et al. – this conference MOPB040 today

Systematic doping vs. EP (FG and LG)

Systematic doping vs. EP (FG and LG)

Palczewski et al. – this conference MOPB039 today

Fitting results remove environmental effects

Sample analysis High resolution – from previous slide

H. Tian et al. – this conference MOPB107 today

The near surface N concentration of Nb sample

with 20 minutes exposure of N2 (20N) was

measured by SIMS with incremental EP.

After 20 µm removal by EP, the concentration

of N keeps constant , but decreases by 60%

after 40 µm.

Pashupati Dhakal et. al ASC’14 1LOr1B-05

Effectiveness of doping vs niobium RRR

No significant signature of doping @ 2.0K on low RRR FG cavities These are 1.5Ghz @ 2.0K

Questions?

Contributing Data C. E. Reece Pashupati Dhakal Grigory Eremeev Hui Tian R.-L. Geng

Funding Special thanks to LCLS-II, especially for taking chance on Nitrogen doping -Initial single cell studies and all 9 cell work

Making nitrogen doping work is not that easy but can be done – off to the cryomodules.

Questions?

Contributing Data C. E. Reece Pashupati Dhakal Grigory Eremeev Hui Tian R.-L. Geng

Funding Special thanks to LCLS-II, especially for taking chance on Nitrogen doping -Initial single cell studies and all 9 cell work

Making nitrogen doping work is not that easy but can be done – off to the cryomodules.

History can come back to bite you

Backup

Furnace doping - redoping

Re-doping with external chemistry changes total absorption rate